Complementary DNA clones coding for the human secreted carbonic anhydrase isozyme (CA VI) have been isolated and their nucleotide sequences determined. These clones identify a 1.45-kb mRNA that is present in high levels in parotid submandibular salivary glands but absent in other tissues such as the sublingual gland, kidney, liver, and prostate gland. Hybridization histochemistry of human salivary glands shows mRNA for CA VI located in the acinar cells of these glands. The cDNA clones encode a protein of 308 amino acids that includes a 17 amino acid leader sequence typical of secreted proteins. The mature protein has 291 amino acids compared to 259 or 260 for the cytoplasmic isozymes, with most of the extra amino acids present as a carboxyl terminal extension. In comparison, sheep CA VI has a 45 amino acid extension [Fernley, R. T., Wright, R. D., & Coghlan, J. P. (1988b) Biochemistry 27, 2815]. Overall the human CA VI protein has a sequence identity of 35% with human CA II, while residues involved in the active site of the enzymes have been conserved. The human sheep secreted carbonic anhydrases have a sequence identity of 72%. This includes the two cysteine residues that are known to be involved in an intramolecular disulfide bond in the sheep CA VI. The enzyme is known to be glycosylated and three potential N-glycosylation sites (Asn-X-Thr/Ser) have been identified. Two of these are known to be glycosylated in sheep CA VI. Southern analysis of human DNA indicates that there is only one gene coding for CA VI.

The secreted carbonic anhydrases, CA VI, are high molecular mass, oligomeric enzymes originally found in the sheep parotid gland and saliva. The enzymes have been purified from the saliva or parotid glands of several different species. All the CA VI enzymes studied have an apparent subunit Mr of about 45,000 as previously reported for the sheep enzyme. By Western analysis, CA VI from human, cow and dog cross-reacted with antibody raised against the purified sheep enzyme whereas that of the mouse did not. The N-terminal sequences of the sheep, human, cow and mouse enzymes are reported. The sheep, cow and human N-terminal sequences are similar to one another while the mouse sequence is substantially different. Nevertheless, the amino acids in the aromatic cluster I (Trp-5, Tyr-7, Trp-16 and Tyr/Phe-20) have all been conserved, as is the case with the cytoplasmic carbonic anhydrases. Eighteen tissues from the sheep have been examined for the presence of CA VI by Western analysis but it has been found only in the salivary glands. Northern analysis and hybridization histochemistry show that the mRNA for CA VI in sheep is expressed specifically in the acinar cells of the parotid and submandibular glands.

In this review we have used our own recent work as a flagship to illustrate the recent renaissance of interest in hybridization histochemistry. A trickle of papers followed the initial key excursion into the in situ labeling of tissue sections (48-50). Our own entry into this field started in 1978 and since then a confluence of important questions and technical advances has served to make hybridization histochemistry much more attractive as a research tool. Hybridization histochemistry is able to solve some problems for which there is no other suitable technique at this time. Hybridization histochemistry provides the location of anatomical sites of gene expression, and viral replication, with uniquely high specificity. We have taken 32P-labeled probes to what appears to be their limit of resolution, which is single cells in thin sections. While 32P has clear disadvantages, exposure time is relatively short and the use of fast-X-ray film to preview the results and estimate exposure time for emulsion has been turned to advantage. Our introduction (27) of the use of whole-mouse sections in hybridization histochemistry has great potential in hormonal, enzymatic, and growth factor gene expression and will no doubt prove of great use in developmental studies and examination of viral infection. The use of synthetic DNA (synthetic oligonucleotides) unshackles the technique from the need for an associated molecular biology laboratory and at once widens the horizon of application of the technique. Although hybridization histochemistry is a valuable research tool which will soon find a niche in many fields, in a short time it should become a key diagnostic aid. It may well become the method of preference for detection of the expression of oncogenes and other cancer-related genes and for viruses which for other reasons are difficult to detect.

No abstract